Academic
Publications
A mesh reconstruction algorithm driven by an intrinsic property of a point cloud

A mesh reconstruction algorithm driven by an intrinsic property of a point cloud,10.1016/S0010-4485(03)00064-2,Computer-aided Design,Hong-wei Lin,Chie

A mesh reconstruction algorithm driven by an intrinsic property of a point cloud   (Citations: 20)
BibTex | RIS | RefWorks Download
Abstract This paper presents an algorithm for reconstructing a triangle mesh surface from a given point cloud. Starting with a seed triangle, the algorithm grows a partially reconstructed triangle mesh by selecting a new point based on an intrinsic property of the point cloud, namely, the sampling,uniformity degree. The reconstructed mesh,is essentially an approximate,minimum-weight,triangulation to the point cloud constrained to be on a two-dimensional manifold. Thus, the reconstructed surface has only small topological difference from the surface of the sampled,object. Topological correct reconstruction can be guaranteed by adding a post-processing step. q 2003 Elsevier Ltd. All rights reserved. Keywords: Point cloud; Intrinsic property; Reconstruction
Journal: Computer-aided Design - CAD , vol. 36, no. 1, pp. 1-9, 2004
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...2004) can be implemented considering the intrinsic nature of the point cloud...

    Yihua Dinget al. An automatic method of measuring foot girths for custom footwear using...

    • ...The relationship between point and point should be clear [2]...

    Meiqing Liuet al. Three-Dimensional Reverse Engineering Modeling and Numerical Simulatio...

    • ...Lin et al. [12] presented an improved method based on an intrinsic property of the point set and tried to overcome the limitation of user-specified parameters...

    Ji Maet al. Delaunay-based triangular surface reconstruction from points via Umbre...

    • ...The mesh reconstruction algorithms can be classified into four categories, namely, implicit surface approach, surface deformation approach, spatial subdivision approach and region-growing approaches [1,2,3] . The first two kind...
    • ...mesh with information of the projected κ neighbor points and led to two unreasonable results where the reconstructed mesh heavily depends on the user-specified parameter κ. Lin et al. [2] searched new...
    • ...In the first mesh growing phase, the triangle candidate is constructed for each non-inner edge e by selecting the best point from points which (1) belonging to the common k neighbor points of the two endpoints of e as well as (2) lying within the search region [2] of e. A...
    • ...The general criteria adopted to select the best point are minimal area criterion, minimal length criterion and weighted minimal length criterion [2] . The...
    • ...In order to further improve the quality of reconstructed mesh, a reformative weighted minimal length rule is presented here inspired by Lin [2] and Eck [10] . The weighted minimal length criterion was presented by Lin [2] , from the analysis of Eck et al. [10] , the precondition of weighted minimal length criterion [2] is that the new...
    • ...In order to further improve the quality of reconstructed mesh, a reformative weighted minimal length rule is presented here inspired by Lin [2] and Eck [10] . The weighted minimal length criterion was presented by Lin [2] , from the analysis of Eck et al. [10] , the precondition of weighted minimal length criterion [2] is that the new...
    • ...In order to further improve the quality of reconstructed mesh, a reformative weighted minimal length rule is presented here inspired by Lin [2] and Eck [10] . The weighted minimal length criterion was presented by Lin [2] , from the analysis of Eck et al. [10] , the precondition of weighted minimal length criterion [2] is that the new...
    • ...difference comparison of different criteria is shown in table 1, where Nin is the point numbers in the point cloud, Nout the vertex numbers in the reconstructed mesh, Ntri the triangle numbers in M, D is the topological differences calculated with the method presented by Lin et al. [2] . From the above comparison, it's evident that our...

    Hanming Lvet al. A Heuristic Approach to Reconstruct Triangle Mesh from Unorganized Poi...

    • ...Starting with a seed triangle, the algorithm of Lin et al. [7] grows a partially reconstructed triangle mesh by selecting a new point based on an intrinsic property of the scattered point cloud, namely, the sampling uniformity degree...
    • ...Our paper presents improvements on existing intrinsic property driven (IPD) algorithm of [7]...
    • ...The rest of our paper is organized as follows: related definitions and point cloud preprocessing are proposed in section 2, improvements on method [7] are described in section 3, experimental results are presented in section 4, and a brief conclusion is given in section 5...
    • ...minimum, therefore point R can satisfy with two principles of “maximum angle” and “minimum length” ,w hile in Reference [7] only the principle of “minimum length” with...
    • ...Reference [7] defined the sampling uniformity degree at a point as the ratio of the lengths of the longest edge and shortest edge incident to the point, and used it to define the close influence region for one active edge...
    • ... best active point cannot be found and a small hole will be generated; (2) With the determination of new neighbor edges, the sampling uniformity degrees for the vertices of a boundary edge may increase, and the new found best active point may improperly lie in the opposite side of the 3D model with sharp angles; (3) The sampling uniformity degrees for one active edge can vary with the adding of new neighbor edges, but in IPD algorithm [7] ...
    • ...Different with method [7], we use the open influence region defined by three faces F1, F2 and F3, as shown in Figure 1. Angle θ is used to control the open influence region, and each face for the region can be represented by one point in the face and the normal vector of the face, as , ( ))...
    • ...Thus, IPD algorithm [7] applies the weighted energy function...
    • ...Our approach is also compared with IPD algorithm [7] on the model of Bunny with 35944 points...
    • ...As shown in Figure 4, the first column are point cloud of bunny and points of its bottom, the second column are recovered surfaces from IPD algorithm [7], the third column are recovered surfaces from our method...
    • ...Based on the existing algorithm [7], three improvements are proposed in our paper including new approaches to determine the seed triangle, define the influence region for the active edge, and then select the best active point...
    • ...Compared with IPD algorithm [7], our approach has the ability to reconstruct the surface preserving more accurate topological coherence with the point cloud...

    Chengjiang Longet al. Improvements on IPD Algorithm for Triangular Mesh Reconstruction from ...

Sort by: